1. Field
The disclosure relates to a method, system, and article of manufacture for a prioritization mechanism for deletion of chunks of deduplicated data objects.
2. Background
Data deduplication is a technique for eliminating redundant data. In a deduplication process, chunks of data are identified and stored during a process of analysis, where the chunks of data are byte patterns. As the analysis continues, other chunks are compared to the stored chunks and whenever a match occurs, the redundant chunk is replaced with a reference that points to a stored chunk. In certain situations the same byte pattern may occur numerous times, and the amount of data to be stored may be greatly reduced by replacing redundant chunks with references that point to at least one unique chunk.
In a deduplication process, a deduplicated data object is broken into many unique chunks. The unique chunks are linked to other base chunks, if the unique chunks are redundant. Otherwise, the unique chunks are stored with associated storage as the unique chunks may become a base chunk.